https://github.com/charlescatta/behavioral-cloning

Convolutional Neural network that drives a car in a simulator
https://github.com/charlescatta/behavioral-cloning

cnn deep-learning keras self-driving-car

Last synced: 2 days ago
JSON representation

Convolutional Neural network that drives a car in a simulator

• Host: GitHub
• URL: https://github.com/charlescatta/behavioral-cloning
• Owner: charlescatta
• Created: 2017-07-31T11:52:05.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-08-19T20:40:44.000Z (about 7 years ago)
• Last Synced: 2023-12-21T18:35:41.457Z (11 months ago)
• Topics: cnn, deep-learning, keras, self-driving-car
• Language: Jupyter Notebook
• Homepage:
• Size: 53.7 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Behavioral Cloning 🚔

[![Udacity - Self-Driving Car NanoDegree](https://s3.amazonaws.com/udacity-sdc/github/shield-carnd.svg)](http://www.udacity.com/drive) [![Docker Automated build](https://img.shields.io/docker/automated/madhorse/behavioral-cloning.svg)](https://hub.docker.com/r/madhorse/behavioral-cloning/)



   



This project uses a _Convolutional Neural Network_ to attempt to learn how to drive a car in a simulator by trying to replicate the driving behaviour of a human player. 

The Neural Net is fed three image streams from cameras fixed on the car and the current steering angle during training.

After training the model is able to send appropriate steering angles to the car in order for it to stay on the track.

----

### Simulator

The car simulator used to gather training data is made by Udacity for their [Self-Driving Car Nanodegree](https://www.udacity.com/drive) program, download it here:

   [MacOS](https://d17h27t6h515a5.cloudfront.net/topher/2017/February/58983385_beta-simulator-mac/beta-simulator-mac.zip)   [Windows](https://d17h27t6h515a5.cloudfront.net/topher/2017/February/58983318_beta-simulator-windows/beta-simulator-windows.zip)   [Linux](https://d17h27t6h515a5.cloudfront.net/topher/2017/February/58983558_beta-simulator-linux/beta-simulator-linux.zip) 

### Running the neural network

To run the neural net, use [docker](https://store.docker.com/search?type=edition&offering=community)

```sh

docker run -p 4567:4567 -it --rm -v `pwd`:/src madhorse/behavioral-cloning python3 drive.py model.h5

```

open your simulator and go in Autonomous Mode, this allows the neural net to recieve images and send steering angles.

### Running training

To run training on the model, use [nvidia-docker](https://github.com/NVIDIA/nvidia-docker) in order to train on the GPU,

use the following commands:

```sh

git clone https://github.com/Charles-Catta/Behavioral-Cloning.git

cd Behavioral-Cloning

wget https://d17h27t6h515a5.cloudfront.net/topher/2016/December/584f6edd_data/data.zip

unzip data.zip

rm data.zip

nvidia-docker run -it --rm -v `pwd`:/src madhorse/behavioral-cloning python3 model.py

```

### Model Architecture

![Model Architecture](img/model.png)

The model architecture for this project is based on Nvidia's paper on [_End to end learning for self-driving cars_](http://images.nvidia.com/content/tegra/automotive/images/2016/solutions/pdf/end-to-end-dl-using-px.pdf)

All of the data preprocessing steps are outlined in the [Jupyter notebook](https://nbviewer.jupyter.org/github/Charles-Catta/Behavioral-Cloning/blob/master/Behavioral_Cloning.ipynb)

Read the [writeup](https://htmlpreview.github.io/?https://github.com/Charles-Catta/Behavioral-Cloning/blob/master/writeup.html)